摘要
为了选择一种既能消除近红外图像噪声又能达到保持其图像边缘要求的消噪方法,通过比较各种常见的图像去噪算法,采用信噪比(SNR)、均方根误差(RMSE)和图像灰度曲面图等作为图像去噪效果的评估。对实际拍摄的多幅近红外图像进行消噪效果对比,仿真实验结果表明:在综合考滤图像去噪平滑效果、图像清晰程度和时间复杂度的基础上,分形消噪法较优,可应用于近红外图像的消噪处理。
In order to improve quality of the near-infrared image with low contrast and the high background noise, the common de-noising algorithms were compared. To get a better de-noising method, which not only removes the near-infrared image noise but also maintain target information, the image signal to noise ratio （SNR） and the root-mean-square error （RMSE） and the image gray surface chart were applied to estimate the de-noising effect of the near-infrared images. A lot of simulations comparing experiments were applied to remove the actual near-infrared image noises with the common de-noising algorithms. The experimental results indicate that the method of the fractal technology removing image noise is superior to others after balancing image smoothness, clearness and time complexity. It can remove effectively noise of the real near-infrared images.
出处
《红外技术》
CSCD
北大核心
2007年第11期657-661,共5页
Infrared Technology
基金
国家自然科学基金项目(项目编号:60575020)
湖北省重点学科黄石理工学院机械电子工程学科建设资助项目
关键词
近红外图像
去噪
小波分析
形态学滤波
分形技术
Near-infrared image
De-noising
Wavelet analysis
Morphology filter
Fractal technology
